Iterative receiver and iterative decoder
An iterative receiver and an iterative decoder are provided. The iterative receiver includes: an external decoder to decode the detected received signal; an impurity adder to add an impurity to an output signal of the external decoder; an interleaver to perform interleaving between the internal detector and the external decoder; and a de-interleaver to perform de-interleaving between the internal detector and the external decoder.
Latest Electronics and Telecommunications Research Institute Patents:
- METHOD AND APPARATUS FOR IMAGE ENCODING, AND METHOD AND APPARATUS FOR IMAGE DECODING
- METHOD AND APPARATUS FOR CONTROLLING MEASUREMENT REPORTS IN COMMUNICATION NETWORK INCLUDING INTEGRATED ACCESS AND BACKHAUL NODE
- METHOD AND APPARATUS FOR ENCODING AND DECODING USING SELECTIVE INFORMATION SHARING BETWEEN CHANNELS BETWEEN CHANNELS
- MPI LAYER GEOMETRY GENERATION METHOD USING PIXEL RAY CROSSING
- METHOD AND APPARATUS FOR TRANSMITTING/RECEIVING SIGNAL BY USING VARIABLE BAND WIDTH IN COMMUNICATION SYSTEM
This application is a 35 U.S.C. §371 national stage filing of PCT Application No. PCT/KR2008/006957 filed on Nov. 26, 2008, which claims priority to, and the benefit of, Korean Patent Application No. 10-2007-0126979 filed on Dec 7, 2007. The contents of the aforementioned applications are hereby incorporated by reference.
TECHNICAL FIELDThe present invention relates to an iterative receiver and an iterative decoder in a communication system, and more particularly, to a data receiving apparatus of a communication system that can use an orthogonal frequency division multiplexing (OFDM) scheme.
This work was supported by the IT R&D program of MIC/IITA. [2006-S-001-02, Development of Adaptive Radio Access and Transmission Technologies for 4th Generation Mobile Communications]
BACKGROUND ARTDue to characteristic of a radio channel, when a digital mobile communication system transmits data, a bit error may occur by various types of causes. Therefore, a channel code that is used to correct the bit error is significantly important.
A turbo code and a low-density parity-check (LDPC) code are generally used as a channel code that approaches a Shannon capacity in an aspect of the configurable decoding complexity. The turbo code and the LDPC code are based on a turbo principle.
The turbo principle denotes iterative exchanging of soft-decision information between a plurality of modules. The turbo principle may iteratively update a log-likelihood ratio (LLR) value of a bit based on an extrinsic LLR value of a decoder. In a process of decoding a channel according to the turbo principle, as a number of iterative decodings increases, a convergence rate may also increase. Therefore, performance may be improved. However, when the number of iterative decodings increases, a calculation amount may also increase. In this case, there may be some problems in improving the performance.
Accordingly, there is a need for an apparatus that can improve iterative receiving performance, iterative decoding performance, and a convergence rate with using an existing turbo principle.
DISCLOSURE OF INVENTION Technical ProblemAn aspect of the present invention provides an iterative receiver and an iterative decoder that can improve iterative receiving performance, iterative decoding performance, a convergence rate by using a turbo code or a low-density parity-check (LDPC) code as an external decoder and applying a modified turbo principle that appropriately adds an impurity to an output signal based on a turbo principle.
Another aspect of the present invention also provides an iterative receiver and an iterative decoder that can significantly improve iterative receiving performance, iterative decoding performance, and a convergence rate by using, as an internal detector, an optimal detector such as a soft interference cancellation with minimum mean square error (SIC-MMSE) detector or a maximum a posteriori (MAP) detector.
Technical SolutionAccording to an aspect of the present invention, there is provided a iterative receiver including: an internal detector to detect a received signal; an external decoder to decode the detected received signal; an impurity adder to add an impurity to an output signal of the external decoder; an interleaver to perform interleaving between the internal detector and the external decoder; and a de-interleaver to perform de-interleaving between the internal detector and the external decoder.
The internal detector may be a soft interference cancellation with minimum mean square error (SIC-MMSE) detector. Also, the internal detector may be a maximum a posteriori (MAP) detector or a suboptimal MAP detector with low complexity.
The impurity adder may add the impurity to extrinsic information of the external decoder, the impurity in which a weight is applied to a priori information of the external decoder.
According to another aspect of the present invention, there is provided an iterative decoder including: an internal decoder to decode a received signal; an external decoder to decode the decoded received signal; an impurity adder to add an impurity to an output signal of the external decoder; an interleaver to perform interleaving between the internal decoder and the external decoder; and a de-interleaver to perform interleaving between the internal decoder and the external decoder.
The impurity adder may adjust the weight to control an amount of impurity to be added to the extrinsic information of the external decoder.
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout. The embodiments are described below in order to explain the present invention by referring to the figures.
A MIMO technology in a radio communication denotes a technology that can transmit signals to a transmitting/receiving end via two or more antennas. As shown in
In the MIMO system, a received signal of an Nt dimension may be represented as,
y=Hx+n. [Equation 1]
H=[h1, h2, . . . , hN
is an
Nr×Nt
channel matrix.
x=[x1, x2, . . . , xN
s is a transmission signal vector. n is an additive white Gaussian noise (AWGN) and may be represented as
n=└n1, n2, . . . , nn
Specifically, x denotes a vector of a transmission signal that is transmitted by a transmitter 101. Each subscript corresponds to two or at least two antennas installed in the transmitter 101. Also, the energy of the transmission signal vector is
E[xxH]=γI.
A signal-to-noise ratio (SNR) for each receiving antenna with respect to
E[nnH]=I
is
γ.
In the MIMO system, a receiver 102 may include an internal decoder or an internal detector 103, a de-interleaver 104, an external decoder 105, and an interleaver 106. When the internal decoder 103 is provided, the receiver 102 may function as an iterative decoder. When the internal detector 103 is provided, the receiver 102 may function as an iterative receiver. Hereinafter, a configuration of the receiver 102 when the internal detector 103 is provided to detect a received signal will be described with reference to
The internal detector 103 may detect the received signal that is transmitted by the transmitter 101 via at least two antennas in the MIMO system. The external decoder 105 may decode the detected received signal.
The de-interleaver 104 may perform de-interleaving between the internal detector 103 and the external decoder 105. Referring to
The interleaver 106 may perform interleaving between the internal detector 103 and the external detector 105. Specifically, referring to
As shown in
Hereinafter, when the internal detector 103 is a maximum a posterior (MAP) detector will be described. The MAP detector is an optimal detector, whereas a calculation amount is extremely great. Therefore, when a relatively large number of antennas and a number of bits for each symbol exist, it may be impossible to configure the MAP detector. However, the MAP detector has excellent iterative receiving performance, and thus can be a model for other detectors.
Referring to
Xj,k
denotes a kth bit of a transmission symbol of a jth transmitting antenna. Mc denotes the number of bits for each symbol.
Xj,kb
is a set of x where
xj,k=b,bε{+1,−1}
and may have
2N
cases.
X[j,k]
denotes a vector that includes an element to satisfy
Xj,k=0.
LA=[LA(x1,1), . . . , LA(xN
denotes a vector that includes priori LLRs.
In the iterative receiver 200, an internal detector 201 may be a soft interference cancellation with minimum mean square error (SIC-MMSE) detector. The internal detector 201 may be a MAP detector or a suboptimal MAP detector with low complexity.
An important characteristic of a conventional turbo principle is an iterative scheme of transmitting and receiving extrinsic information between two internal detectors, or between an internal detector and an external detector to thereby converge the extrinsic information to a global optima. It is possible to maintain statistical independence between an internal element and an external element by applying the iterative scheme. In comparison to
Extrinsic information corresponds to information excluding a priori information from a posteriori information. According to the present invention, the impurity adder 205 may add a portion of a priori information to the extrinsic information instead of excluding all the a priori information from the a posteriori information.
Specifically, the impurity adder 205 may add an impurity to extrinsic information of an external decoder 203. In the impurity, a weight is applied to a priori information of the external decoder 203. The impurity adder 205 may adjust the weight to control an amount of impurity to be added to the extrinsic information of the external decoder 203.
The effect that can be obtained by adding the impurity to the extrinsic information of the external decoder 203 will be described later with reference to
For example, the external decoder 203 may decode a received signal detected by the internal detector 201, or a received signal decoded by the internal decoder 201, using a turbo code based on a turbo principle, or using an LDPC code based on the turbo principle. The impurity adder 205 may apply a weight to a priori information of the external decoder 203 to adjust an amount of impurity.
The interleaver 204 may interleave a result that is obtained by adding to extrinsic information of the external decoder 206 with the impurity in which the weight is applied to a priori information corresponding to an input signal of the external decoder 203.
A prior LLR of the internal detector 201 may be calculated as given by,
LA
denotes the priori LLR of the internal detector 201.
LP
denotes a posteriori information by the external decoder 203.
LE
denotes an extrinsic LLR of the external decoder 203 when the impurity is not added.
LIE
denotes the extrinsic LLR in which the impurity is added by the impurity adder 205.
LA
denotes a pure priori LLR by the external decoder 203.
LA
is a priori LLR in which the impurity is added and also denotes the impurity in
π
denotes interleaving by the interleaver 204
β.
denotes the weight to adjust the amount of impurity. When
β=1,
it may denote a conventional iterative receiver. It is possible to design a system to appropriately set the weight
β.
Specifically, in comparison to
In Equation 4, a prior information
LA
of the internal detector 201 denotes a result that is obtained by interleaving extrinsic information.
LIE
The extrinsic information
LIE
is obtained by subtracting a posteriori information
LP
by the external decoder 203 from the multiplication result between a priori information
LA
of the external decoder 203 and the impurity).
β.
Specifically, the impurity adder 205 denotes adding the impurity in which a predetermined weight, for example,
1−β,
is applied to a priori information of the external decoder 203, to the pure extrinsic information
LE
of the external decoder 203.
The iterative receiver 300 may further include a de-mapper 302 and a soft input/output (I/O) decoder 304 in addition to the iterative receiver 200 of
The SIC-MMSE detector 301 corresponding to a non-optimal internal detector may relieve a remaining interference of a soft-interference canceled signal via an MMSE filter, which may be expressed by the following Equation 6. A jth antenna symbol after soft interference cancellation may be given by:
yj=y−H
denotes a symbol vector estimated by a symbol estimator 306. The SC-MMSE detector 301 may filter a signal
yj
in which interference is removed to thereby generate a signal
Ŝj.
The de-mapper 302 may de-map the generated signal
Ŝj
to thereby generate an existing signal. The conditional probability of the signal
Ŝj
according to Gaussian approximation may be expressed as,
μj
denotes the average of
Ŝj
that is generally obtained according to the Gaussian approximation and
vj2
is a variance of
Ŝj.
Therefore, an extrinsic LLR calculated by the SIC-MMSE detector 301 may be expressed as,
Sj+b
is a set of x where
xj,k=b,bε{+1,−1}
and has a
2N
number of cases.
The EXIT graph is an analytical instrument to form extrinsic information transmitted and received between an internal detector and an external detector into an iterative performance orbit. A curve MAP (maxlog) indicates an approximate value of an exact MAP detector according to a maxlog approximation scheme.
Referring to
β=0.5
and when
β=0.0,
a contacting area between the SC-MMSE detector and the turbo decoder increases.
Accordingly, as a result of iterative performance, SC-MMSE detector is possible to receive a signal in which an error is excluded, like the MAP detector. This result may be obtained since mutual information of the contacting area increases due to the impurity added to the extrinsic information.
Specifically, when an extrinsic LLR of a decoder is greater than or equal to a predetermined level to be reliable and in this instance, a priori LLR is partially added to a pure extrinsic LLR, it is possible to improve the reliability of the extrinsic LLR that is output by the detector. The extrinsic LLR of the detector with the improved reliability may improve the reliability of the extrinsic LLR of the decoder.
In comparison to the conventional turbo principle, in the case of the modified turbo principle according to the present invention, side effect of interfering statistical independence between the detector and the decoder may be caused due to the impurity. Specifically, as shown in
However, the above side effect may be neglected in comparison to a pure effect by the impurity and may also be solved by an appropriate selection of a weight. For example, the SIC-MMSE detector may show excellent tradeoff between the pure effect and the side effect as given by,
β1=φi−1 [Equation 9]
i indicates an iterative sequence.
For example,
Referring to
Also, when using a MAP (proposed) detector based on the modified turbo principle according to the present invention, the performance was improved by about 0.2 dB in comparison to the iterative receiver based on the conventional turbo principle. For example, the EXIT graph of
As induced from
According to the present invention, there are provided an iterative receiver and an iterative decoder that can partially add a priori LLR to a pure extrinsic LLR and thereby increase reliability of an extrinsic LLR that is output by a detector.
Also, according to the present invention, there are provided an iterative receiver and an iterative decoder that can significantly improve iterative receiving performance, iterative decoding performance, and a convergence rate by using, as an internal detector, an optimal detector such as an SIC-MMSE detector or a MAP detector.
Also, according to the present invention, there are provided an iterative receiver and an iterative decoder that can improve iterative receiving performance, iterative decoding performance, a convergence rate by using a turbo code or an LDPC code as an external decoder and applying a modified turbo principle that appropriately adds an impurity to an output signal based on a turbo principle.
Although a few embodiments of the present invention have been shown and described, the present invention is not limited to the described embodiments. Instead, it would be appreciated by those skilled in the art that changes may be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims
1. An iterative receiver comprising:
- an internal detector to detect a received signal;
- an external decoder to decode the detected received signal;
- an impurity adder to add an impurity to an output signal of the external decoder;
- an interleaver to perform interleaving between the internal detector and the external decoder; and
- a de-interleaver to perform de-interleaving between the internal detector and the external decoder.
2. The iterative receiver of claim 1, wherein the internal detector is a soft interference cancellation with minimum mean square error (SIC-MMSE) detector.
3. The iterative receiver of claim 1, wherein the internal detector is a maximum a posteriori (MAP) detector or a suboptimal MAP detector with low complexity.
4. The iterative receiver of claim 1, wherein the external decoder decodes the detected received signal using a turbo code.
5. The iterative receiver of claim 1, wherein the external decoder decodes the detected received signal using a low-density parity-check (LDPC) code.
6. The iterative receiver of claim 1, wherein the impurity adder adds the impurity to extrinsic information of the external decoder, the impurity in which a weight is applied to a priori information of the external decoder.
7. The iterative receiver of claim 6, wherein the impurity adder adjusts the weight to control an amount of impurity to be added to the extrinsic information of the external decoder.
8. An iterative decoder comprising:
- an internal decoder to decode a received signal;
- an external decoder to decode the decoded received signal;
- an impurity adder to add an impurity to an output signal of the external decoder;
- an interleaver to perform interleaving between the internal decoder and the external decoder; and
- a de-interleaver to perform interleaving between the internal decoder and the external decoder.
9. The iterative decoder of claim 8, wherein the impurity adder adds the impurity to extrinsic information of the external decoder, the impurity in which a weight is applied to a priori information of the external decoder.
10. The iterative decoder of claim 8, wherein the impurity adder adjusts the weight to control an amount of impurity to be added to the extrinsic information of the external decoder.
7434145 | October 7, 2008 | Jin et al. |
8059729 | November 15, 2011 | Zhidkov |
20030014714 | January 16, 2003 | Yamada |
20040062302 | April 1, 2004 | Fujii et al. |
20050235195 | October 20, 2005 | Choi et al. |
20070136648 | June 14, 2007 | Kwon et al. |
1020000076753 | December 2000 | KR |
1020060118313 | November 2006 | KR |
- Douillard, Catherine et al., “Iterative Correction of Intersymbol Interference: Turbo-Equalization,” European Transactions on Telecommunications, vol. 6(5):507-511 (1995).
- Haykin, Simon et al., “Turbo-MIMO for Wireless Communications,” IEEE Communications Magazine, vol. 42(10):48-53 (2004).
- Witzke, Melanie et al., “Iterative Detection of MIMO Signals with Linear Detectors,” Signals, Systems and Computers, vol. 1:289-293 (2002).
Type: Grant
Filed: Nov 26, 2008
Date of Patent: Dec 4, 2012
Patent Publication Number: 20100266069
Assignee: Electronics and Telecommunications Research Institute (Daejeon)
Inventors: Junyoung Nam (Daejeon), Seong Rag Kim (Daejeon), Hyun Kyu Chung (Daejeon)
Primary Examiner: Kevin Kim
Attorney: Nelson Mullins Riley & Scarborough LLP
Application Number: 12/746,744
International Classification: H04L 27/06 (20060101); H03M 13/03 (20060101);